01; 
the 
sor 
"tic 
d 6 
sa 
wo 
lor 
red 
ely 
ind 
ave 
ase 
or 
ion 
dku 
ern 
the 
es. 
Sic 
the 
een 
id 6 
om 
the 
t 1s 
ing 
lon 
ted 
> of 
| of 
ace 
xed 
Ires 
lof 
ere 
ital 
on: 
| ST- 
ixel 
the 
IM 
T - 
International Archives of the Photogrammetry, Remote Sensing 
The spectral radiances L lambda were converted into effective 
satellite temperatures T by 
T=K2/In{ Kl /L +1) (2) 
where K1, K2 are calibration constants. By applying this 
equation on the two types of TM sensors, we calculated the 
following; 
For Landsat TM $ the constants are K1— 60,776 and K2 
1260,56 [mW em-2 sr-1 um-1] 
For Landsat TM 7 the constants are K 1— 666,09 and K2 
1282.71 [mW cm-2 sr-1 um-1] 
Il 
il 
3, Results and Discussions 
Figure (2) shows the enhanced images in both dates for 
water color discrimination. The yellowish brown color groups 
together all the plumes due to freshwater discharges entering the 
sea. This color indicates the presence of organic compounds 
with sufficient concentrations in these freshwater plumes. The 
dark blue color) for clear water (case 1 water), where clear 
water absorbs relatively little energy having wavelengths less 
than 0.6 um. High transmittance typifies these wavelengths with 
a maximum in the blue-green portion of the spectrum. The 
brown color at the inlet of Abu kir., the reflectance of water 
changes with the chlorophyll concentration involved. Increases 
in chlorophyll concentration tend to decrease water reflectance 
in blue wavelengths and increase it in green wavelengths. 
Six profiles were drawn for the two enhanced images in 
figure (3). It is obvious that the clear water profile sample (1) 
shows maximum reflectance in band 1 and least reflectance in 
band 3 and appear in blue color in the image as the blue (0.45 to 
0.52 um) band 1 provides increased penetration of water bodies, 
The shorter-wave length cutoff is just below the peak 
transmittance of clear water, Wavelengths below 0.45 um are 
substantially influenced by atmospheric scattering and 
absorption. Also, this is due to that the red (0.63 to 0.69 um) 
band 3 chlorophyll absorption band of healthy green vegetation 
that represents one of the most important bands for vegetation 
discrimination. Profile of sample (4) shows least reflectance in 
band 1 and relatively high reflectance in band 2 and 3 as it 
represents Edku outlet where the discharge of agricultural 
wastes of EL-Maadiya channel, this is due to the presence of 
dissolved organic matter (yellow matter) that absorbs very little 
in the red, but its absorption increases rapidly with decreasing 
wavelength and can be significant at blue wavelength. 
Therefore, in coastal water, the yellow matter is the dominant 
absorber in the blue end of the spectrum. Profiles of samples 
(2,3,5,6) show relatively low reflectance in band 1 and higher 
reflectance in both bands 2 and 3 due to urban activities 
influence on sample no 2, thermal pollution of el-tabia pumping 
station at sample no 3, organic particles due to sewage 
discharge at sample no 5 and oil pollution resulting from 
shipping wastes at sample no 6. 
For the temporal analysis, it is obvious that landsat 
classified image for year 2000 shows enhancement in the 
coastal water quality than landsat 1990 due to enforcement of 
environmental law 1994 for preventing the wastewater 
discharges into sea. Temporal reflectance analyses of the 
selected coastal water sites from satellite images TM 1990 and 
2000. Also, Land use maps have been generated for the coastal 
area illustrating the landuse activities along the coast. 
1297 
and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004 
Band 6 for thermal infra-red was used for thematic 
mapping of the coastal water as illustrated in figures (6, 7). 
Simultaneous measurements of temperature of water bodies 
directly at surface 199] and 2000 correspond with the 
calculated temperature in a range of 1 Kelvin by an emissivity 
of water of about 0.98 —0.99. Temperatures were converted 
from degree Kelvin into degree Celsius for simpler handling. 
After calculation of day-night differences and the changes of 
day-night differences between 1991 and 2000 the values were 
rounded to 1? Celsius. It is obvious from the figures that 
the highest temperature detected from thee satellite were at the 
location of discharge of the pollutants. 
  
  
Figure (1) Landuse maps for the Alexandria 
Coast for 1990 and 2000